Skip to content

DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 is served by 29 companies at prices ranging from $0.04 to $0.44 per million input tokens, a 10.1× difference for identical weights.

cheapest host · Wafer
$0.04
price spread across hosts
10×
companies serving it
29
intelligence index
n/a

Is it worth it?

No cheaper model matches it on quality.

It beats GLM 5.2, DeepSeek V4.1 Flash at the same quality band, for more money.

Every host, compared

The same weights, priced and configured differently by 29 companies. Precision matters: 4-bit builds measurably degrade output and cost the same.

Host$/1M in$/1M outCachedContextPrecisionUptime
Wafercheapest$0.043$0.300$0.0351.05Mnvfp4100%
DeepInfra$0.048$0.144$0.0121.05Mfp899.9%
DeepInfra$0.060$0.180$0.0151.05Mfp899.9%
Relace$0.070$0.140$0.0141.05Mfp499.3%
StreamLake$0.088$0.264$0.0031.02Mfp8100%
DeepInfra$0.090$0.270$0.0221.05Mfp899.4%
Reka$0.100$0.420$0.060262Kunknown99.2%
DigitalOcean$0.119$0.238$0.0241.05Munknown100%
Wafer$0.130$0.230$0.1251.05Mnvfp4100%
BaseTen$0.130$0.260$0.0281.05Mfp8100%
BaseTen$0.130$0.260$0.0281.05Mfp32100%
Venice$0.130$0.260$0.0281.00Munknown100%
CoreWeave$0.130$0.270$0.050262Kfp892.2%
Inceptron$0.130$0.495$0.0271.05Mfp4100%
Cohere$0.140$0.280$0.0701.05Mfp899.5%
Parasail$0.140$0.280$0.0501.05Mfp898.7%
Sail Research$0.150$0.300$0.0421.05Mfp4100%
Sail Research$0.150$0.300$0.0421.05Mfp4100%
Mancer 2$0.150$0.500$0.0751.05Mfp899.6%
SiliconFlow$0.220$0.660$0.0281.05Mfp899.8%
GMICloud$0.286$0.858$0.0091.05Mfp8100%
Alibaba$0.352$1.06$0.0351.00Munknown100%
Alibaba$0.374$1.12$0.0371.00Munknown100%
Novita$0.409$1.23$0.0261.05Mfp8100%
Chutes$0.440$1.32$0.0441.05Mfp8n/a
PrimeIntellect$0.440$1.32$0.0141.05Mfp4100%
Baidu$0.440$1.32$0.0141.05Mfp899.9%
AtlasCloud$0.440$1.32$0.0281.05Mfp4100%
Cloudflare$0.440$1.32$0.0141.05Munknown100%

6 of 29 hosts serve this model at 4-bit precision (Relace, Inceptron, Sail Research, Sail Research, PrimeIntellect, AtlasCloud), at the same price as full-precision hosts. See why this matters.

What it costs to run

Marketing emails
$0.161
per 1,000 · short brief in, one email out
Support replies
$0.168
per 1,000 · ticket plus help-centre context
Lead enrichment
$0.325
per 1,000 · company page in, fields out
Long-form drafts
$0.652
per 1,000 · outline in, full draft out
Questions on docs
$0.920
per 1,000 · one large document, many questions
Agent / tool loop
$0.746
per 1,000 · long context reused each step

Priced on the median host. Change the workload on the leaderboard to see how the ranking moves.

Specification

Context, most hosts
1.05M · 1,573 pages
Context range across hosts
262K to 1.05M
Max output
384K
Cache discount
89% off cached input
Input types
text
Tool use
supported
Reasoning
optional
Arena Elo
1185 across 8 categories
Agentic Elo
n/a
AA Intelligence
n/a

Other models with the same host spread

Host pricing and precision read live from OpenRouter’s endpoint API, compiled 16 November 2026. Free to quote with attribution. See all findings.